INDEX

Explanations

terms related to optimization and maximizing efficiency

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

heits

-0.17

apon

-0.17

eled

-0.16

leton

-0.15

iard

-0.15

iban

-0.15

evice

-0.15

ible

-0.15

itan

-0.15

esis

-0.14

POSITIVE LOGITS

ally

0.23

izers

0.22

ised

0.22

izing

0.21

ized

0.21

istic

0.21

izes

0.21

istically

0.20

isation

0.20

ALSE

0.20

Activations Density 0.007%

terms related to optimization and maximizing efficiency

No Comments

No Known Activations